| Sign In to gain access to subscriptions and/or personal tools. |
Influence of Contamination on Zirconia Ceramic Bonding
1 Department of Prosthodontics, Propaedeutics and Dental Materials, Christian-Albrechts University at Kiel, Arnold-Heller-Str. 16, 24105 Kiel, Germany; and Correspondence: * corresponding author, mkern{at}proth.uni-kiel.de
The removal of contaminants prior to the bonding of ceramics is critical for the clinical success of a long-term durable resin bond. This study tested the null hypotheses that there are no contaminants on the zirconia ceramic surface left after try-in simulation, and there are no influences of contamination and cleaning methods on zirconia ceramic bonding durability with 10-methacryloyloxy-decyl dihydrogenphosphate-containing composite resins. After saliva immersion and the use of a silicone disclosing agent, airborne-particle-abraded ceramic specimens were cleaned with acetone, 36% phosphoric acid, additional airborne-particle abrasion, or only water spray. Chemical analyses of specimen surfaces were performed by x-ray photoelectron spectroscopy. The influences of contamination and cleaning methods on ceramic bond durability were examined by tensile testing after 3 or 150 days’ water storage with 37,500 thermal cycles. Contamination, existing after try-in simulation as confirmed by chemical analysis, significantly reduced zirconia ceramic-resin bonds. Airborne-particle abrasion may be the most effective cleaning method.
Key Words: zirconia ceramic bonding • contamination • cleaning
Astrong, durable resin-ceramic bonding provides all-ceramic restorations with high retention, improved marginal adaptation, and increased fracture resistance of the restored teeth and the restorations, although conventional cements can also be used for luting zirconia ceramic restorations, in most cases (Blatz et al., 2003). In particular, 10-methacryloyloxy-decyl dihydrogenphosphate (MDP)-containing composite resins showed a long-term durable bond to zirconia ceramic after airborne particle abrasion (Kern, 2000; Wegner and Kern, 2000; Blatz et al., 2003). However, a strong resin-ceramic bond achieved in strictly controlled clean in vitro tests might be compromised in clinical situations, leading to significantly reduced bond strength. During clinical try-in procedures, contamination of restorative luting surfaces by saliva, blood, or silicone indicators cannot be avoided (van Schalkwyk et al., 2003). Saliva contamination is frequently a main reason for reduced resin bond strength (Aboush, 1998; Bishara et al., 2002; Cacciafesta et al., 2003; van Schalkwyk et al., 2003; Eiriksson et al., 2004). In dental textbooks, organic solutions are recommended for the removal of saliva contamination on luting surfaces of restorations before cementation (Rosenstiel et al., 1995). In the instructions of modern adhesive composite resins, phosphoric acid gel is sometimes recommended for the removal of contaminants. Due to the chemical stability of set silicone, it is believed that a clean, residue-free fitting surface remains after removal of a set silicone indicator film after a try-in procedure (technical instructions of Fit checker, GC Co., Tokyo, Japan). However, the silicone-disclosing procedure may leave a thin layer of residual unpolymerized organic film on the bonding surfaces of the restorations, leading to compromised resin bonding (Millstein et al., 1989; Sorensen, 1991; Szep et al., 2003). Some investigators presumed that chemical reactions (Sheth et al., 1988; Sorensen, 1991) and covalent bonds (Millstein et al., 1989) might occur between silicone indicator films and restorations, leading to a stable adherence of silicone to bonding substrate, and therefore reducing resin bonding. MDP-containing composite resins, e.g., Panavia 21 and Panavia F, showed a long-term durable bond to airborne-particle-abraded zirconia ceramic after water storage for 150 days (Kern and Wegner, 1998; Wegner et al., 2002) and 2 yrs (Wegner and Kern, 2000) with repeated thermal cycling. However, after saliva and silicone contamination and different cleaning methods simulating clinical conditions, the long-term bond to zirconia ceramic with Panavia F 2.0 was not stable (Quaas et al., 2007). In this previous study, the presence of contaminants and the effectiveness of cleaning methods have not been identified by chemical analysis. Therefore, in this study, we used x-ray photoelectron spectroscopy (XPS), a highly surface-sensitive technique for determining the chemical composition of multiphase compounds and for detecting surface contaminants (Ratner and Castner, 1997), to identify the existence of saliva and silicone contamination on zirconia ceramic surfaces after try-in simulation. In addition, the influence of long-term water storage with thermal cycling on bonding durability of MDP-containing composite resins to ceramic after contamination and cleaning was investigated. The null hypotheses tested were that there are no contaminants on ceramic surfaces after try-in simulation, and that there is no influence of contamination and cleaning procedures on zirconia ceramic bonding with MDP-containing composite resins.
Specimen Preparation Densely sintered, partially stabilized zirconia ceramic disk-like specimens (Cercon, DeguDent, Hanau, Germany) were used in this study (Fig. 1  ). The specimen surfaces were wet-polished with 600-grit silicon carbide paper and then airborne-particle-abraded with 50µm Al2O3 at 2.5 bar pressure for 15 sec at a distance of 10 mm. Subsequently, the surfaces were air-cleaned for 20 sec with the Rocatector delta device (3M ESPE, Seefeld, Germany). For XPS examination, specimens were cleaned ultrasonically in distilled water for 10 min. Then, the cleaned specimens were immersed in saliva for 1 min. Saliva was collected from one healthy female donor who had refrained from eating and drinking for 1.5 hrs prior to the collection procedure, which was approved by the ethics committee of Christian-Albrechts University at Kiel. All experiments were performed with fresh saliva collected at the same time. After saliva immersion, the specimens were rinsed with tap water for 15 sec and air-dried for 15 sec. The specimens were then pressed into the mixed silicone indicator (Fit Checker black, GC Corporation, Tokyo, Japan) by finger pressure for 3 min. For XPS examination and TBS testing, the specimens were classified into 5 test groups, i.e., with 4 cleaning methods generally available in dental offices and one control group (Fig. 1): (AC) immersed in acetone for 15 sec and then rinsed with tap water for 15 sec; (HP) etched with 36% phosphoric acid gel (Conditioner 36, Dentsply DeTrey, Konstanz, Germany) for 30 sec two times and then rinsed with tap water for 30 sec; (AA) airborne-particle-abraded with 50 µm Al2O3 at 2.5 bars pressure for 15 sec at a distance of 10 mm and then cleaned with compressed air; (TW) rinsed only with tap water; and (CAA) clean airborne-particle-abraded specimens without contamination as control group.
XPS Examination Specimens of the above test groups and, additionally, specimens immersed only in saliva (SA) were examined with XPS (Vacuum Generators, London, England) for chemical identification of the contaminants and evaluation of the effectiveness of the cleaning methods (Fig. 1  ). All measurements were done with XPS with an x-ray source providing Al K x-rays. A pass energy setting of 100 eV allowed for adequate quantitative analysis, yielding high-resolution scans of the carbon (C1s), oxygen (O1s), zirconium (Zr3d), and silicon (Si2p) peaks.
Tensile Bond Strength (TBS) Testing
SEM Examination
Statistical Analysis
XPS Results The peak intensity ratios of C/O, O/C, and C/Zr in test groups are shown in Table 1 . After saliva contamination, the C/Zr and O/Zr ratios drastically increased, indicating that the ceramic surface was covered with an organic coating, mainly composed of carbon (C) and oxygen (O), compared with the control group. However, after try-in simulation, Si was also found on the ceramic surface, in addition to increased C/Zr and O/Zr ratios. After the ceramic was cleaned with phosphoric acid or by airborne-particle abrasion, ratios for O/Zr and C/Zr were reduced, comparable with those of the uncontaminated control. However, the Si contamination on the surface was only partially removed after phosphoric acid cleaning, in contrast to the complete removal of Si by airborne-particle abrasion. After acetone cleaning, although the concentrations of C and O decreased compared with those of the ’water only’ cleaning, the amounts of C and O at the surface were still considerably higher than in the control group. However, Si seemed to be removed completely by acetone cleaning.
TBS Results Medians, means, and standard deviations of TBS in MPa of the tested groups are shown in Table 2 . For both luting resins, water cleaning showed significantly lower initial TBS than the control (p  0.01), and during storage with thermal cycling, all specimens debonded spontaneously. For both luting resins, cleaning with airborne-particle abrasion resulted in relatively high initial TBS, which remained statistically stable over 150 days’ water storage with thermal cycling, compared with the control. Acetone cleaning did not increase TBS significantly for both luting resins as compared with ’water only’ cleaning (p > 0.05). For most test conditions, there was no statistical difference in TBS between the luting resins PF 2.0 and P21 (p > 0.05). For both luting resins, initial TBS after phosphoric acid cleaning were statistically higher than after water or acetone cleaning (p 0.05), but statistically lower than after air-abrasion or in the control group (p 0.05). After 150 days with thermal cycling, TBS decreased statistically for both luting resins, but the TBS of P21 was statistically higher than that of PF 2.0.
SEM Observation and Fractographic Analysis Mean percentages of areas assigned to the failure modes observed in the bonding groups after tensile bond strength testing are shown in Fig. 2a . In the control group and after cleaning with air-abrasion (Figs. 2b, 2c), failures were found to be mostly cohesive in the luting composite resins (PF 2.0 or P21) (Fig. 2d) or in the tube-filling composite resin (Clearfil FII) (Fig. 2e). In contrast, after acetone or water cleaning, the failure mode was completely adhesive at the zirconia ceramic surface, without any composite resin residue. After phosphoric acid cleaning, the failure mode was mostly cohesive for both luting resins, with a high initial TBS. However, after 150 days, for group PF 2.0 there was a significant increase of adhesive failure area accompanied by TBS lower than that of group P21 (Fig. 2), while the cohesive failure portion in the PF 2.0 luting resin increased.
Factors influencing resin bonding to zirconia ceramic include the wettability of ceramic by adhesive resins, the roughness of ceramic surface, the composition of adhesive resins, the handling performance of adhesive resins, and possible contamination during bonding procedures. After saliva immersion, salivary protein adsorption occurs not only to the tooth surface (Baier and Glantz, 1978), but also on the restorative materials (Hannig, 1997; Carlen et al., 2001). In this study, the hydrophilic ceramic surface showed a roughened "activated" surface after airborne-particle abrasion (Hummel and Kern, 2004). Non-covalent adsorption of salivary proteins occurred easily on this surface during saliva immersion, and could not be removed by water rinsing, as showed by XPS. Silicone contamination is a well-known problem in materials science. The main component of silicone disclosing agent is poly(dimethylsiloxane), containing a Si-O backbone. Organic groups (CH3) can attach via Si-C bonds to this backbone, which may occur during contamination in bonding applications. In this study, the presence of Si on ceramic surfaces after try-in simulation demonstrated a silicone residue on ceramic surfaces after the bulk silicone indicator was peeled from ceramic surfaces. Our TBS results showed that, for both MDP-containing luting resins, it was impossible to achieve a stable bond to ceramic after try-in simulation, while a long-term stable bond to clean zirconia ceramic was achieved in the uncontaminated control group, confirming previous results (Kern, 2000; Wegner and Kern, 2000; Wegner et al., 2002). XPS generally detects photoelectrons approximately 2–10 nm at the tops of specimen surfaces (Moulder et al., 1992). Therefore, there was an ultra-thin layer (less than 10 nm) of salivary and silicone contaminants covering the ceramic surface, since Zr signal originating from the ceramic substrate was still detectable. Therefore, the null hypothesis—that there are no contaminants on a ceramic surface left after try-in simulation—had to be rejected. Among the tested cleaning methods, airborne-particle abrasion was the most effective method of removing contaminants, as shown by XPS and TBS results. These results further confirmed that airborne-particle abrasion not only removed contaminants from ceramic surfaces (Kern and Thompson, 1993, 1994), but also exposed a fresh bonding surface by mechanical removal of superficial ceramic, contributing to a strong durable ceramic bond with MDP-containing composite resins. In a recent study (Quaas et al., 2007), after saliva immersion and the application of silicone-disclosing medium, and after the ceramic was etched twice with phosphoric acid (36%) for 30 sec, TBS was statistically higher than that after the ceramic was etched only once with phosphoric acid for 60 sec. Therefore, in this study, etching twice with phosphoric acid was performed. As shown by XPS, phosphoric acid cleaning seems to be as effective a method of removing salivary contamination as airborne-particle abrasion. However, this cleaning obviously did not completely remove silicone residue from ceramic surface, indicated by the decreased long-term TBS for both luting resins. In addition, the phosphoric acid and water spray might decrease the surface energy of the activated ceramic surface, leading to reduced TBS. Generally, the activated ceramic surface is sensitive to the environment and will partially lose its wettability because of air contamination (Surface preparation for improved adhesion, 2001). Airborne-particle abrasion produced an activated rough ceramic surface, which might have made complete air-drying difficult, probably leading to the moisture contamination of the ceramic surface during water rinsing. Regarding the composition of P21 and PF 2.0, although the basic components are similar, PF 2.0 contains additional photo-initiators and a fluoride compound. The matrix monomer of PF 2.0 has been modified to maintain its mechanical properties after releasing fluoride ions (personal communication with Dr. Kazumitsu Nakatsuka, Kuraray, Osaka, Japan). However, in this study, after phosphoric acid cleaning, the long-term TBS of PF 2.0 was significantly lower than that of P21. These results indicate that PF 2.0 is probably not as stable as P21, or the photo-initiators in PF 2.0 might be more sensitive to moisture resulting from water spray than are the chemical initiators in P21, in agreement with one recent report (Lüthy et al., 2006). Acetone cleaning seems to be effective in removing silicone contamination, but was not effective in removing salivary contaminants, as shown by XPS results and low TBS. This fact indicates that TBS might be more affected by C and O remaining from saliva than by residual Si, which was greatly influenced by different cleaning methods. Therefore, the null hypothesis—that there is no influence of contaminants and cleaning methods on zirconia ceramic bonding with MDP-containing composite resins—had to be rejected. The combination of chemical identification with XPS and long-term TBS was an effective method of determining the effects of contaminants and cleaning methods on zirconia ceramic bonding.
The study was jointly funded by the Department of Prosthodontics, Propaedeutics and Dental Materials and the Multicomponent Materials Institute, Christian-Albrechts University at Kiel. Received for publication September 6, 2006. Revision received March 30, 2007. Accepted for publication April 19, 2007.
Journal of Dental Research, Vol. 86, No. 8,
749-753 (2007) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
